Abstract:This paper focuses on the problem of 6D object pose estimation in occluded environments. We propose a novel 6D pose estimation algorithm based on RGB‐D images, which utilises conventional image processing techniques to generate a distance map and uses it as the compensation information for pose estimation. Specifically, we introduce a new sampling strategy and add an object segmentation module based on foreground probability to improve the algorithm's robustness to occluded environments. 6D object pose estimation usually refers to acquiring the 6D pose information of 3D objects in the sensor coordinate system using computer vision techniques. However, the task faces numerous challenges due to the complexity of natural scenes. One of the most significant challenges is occlusion, which is an unavoidable situation in 3D scenes and poses a significant obstacle in real‐world applications. To tackle this issue, we propose a novel 6D pose estimation algorithm based on RGB‐D images, aiming for enhanced robustness in occluded environments. Our approach follows the basic architecture of keypoint‐based pose estimation algorithms. To better leverage complementary information of RGB‐D data, we introduce a novel foreground probability‐guided sampling strategy at the network's input stage. This strategy mitigates the sampling ratio imbalance between foreground and background points due to smaller foreground objects in occluded environments. Moreover, considering the impact of occlusion on semantic segmentation networks, we introduce a new object segmentation module. This module utilises traditional image processing techniques to compensate for severe semantic segmentation errors of deep learning networks. We evaluate our algorithm using the Occlusion LineMOD public dataset. Experimental results demonstrate that our method is more robust in occlusion environments compared to existing state‐of‐the‐art algorithms. It maintains stable performance even in scenarios with no or low occlusion.

DCNet: Dense Correspondence Neural Network for 6DoF Object Pose Estimation in Occluded Scenes

3D Point-to-Keypoint Voting Network for 6D Pose Estimation

RDPN6D: Residual-based Dense Point-wise Network for 6Dof Object Pose Estimation Based on RGB-D Images

BiCo-Net: Regress Globally, Match Locally for Robust 6D Pose Estimation

KDFNet: Learning Keypoint Distance Field for 6D Object Pose Estimation

RFFCE: Residual Feature Fusion and Confidence Evaluation Network for 6dof Pose Estimation.

Neural Correspondence Field for Object Pose Estimation

DSC-PoseNet: Learning 6DoF Object Pose Estimation via Dual-scale Consistency

Attention Guided 6D Object Pose Estimation with Multi-constraints Voting Network

CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-Dof Object Pose Estimation

DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion

Occluded object 6D pose estimation using foreground probability compensation

RNNPose: 6-DoF Object Pose Estimation Via Recurrent Correspondence Field Estimation and Pose Optimization

SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimation

Deep Learning-Based 6-DoF Object Pose Estimation Considering Synthetic Dataset

6D pose estimation of 3D objects in scenes with mutual similarities and occlusions

PA-Pose: Partial Point Cloud Fusion Based on Reliable Alignment for 6D Pose Tracking

DenseFusion-DA2: End-to-End Pose-Estimation Network Based on RGB-D Sensors and Multi-Channel Attention Mechanisms

A modal fusion network with dual attention mechanism for 6D pose estimation

SilhoNet: An RGB Method for 6D Object Pose Estimation

Exploring Multiple Geometric Representations for 6DoF Object Pose Estimation